What is the easiest way to only display decimals for the float 25.50 in C?

Question

Just starting out C. For instance given the number float $25.50. I want to display the dollars and cents separately. So for dollars I can use %.0f to display 25. How can I only display the 50?

Answer 1

You're probably looking for modf() from <math.h>. That will split the value into the integer part and the fractional part.

double amount = 25.50;
double dollars;
double cents = modf(amount, &dollars);

You can then print just the fraction:

printf("%.2f", cents);

or, if you're not going to need the fractional value again:

printf("%.2f", modf(amount, &dollars));

There isn't a way using just printf() to print only the fractional part of a number; you have to modify the value passed to it for printing so that it only sees the fractional part.

(Note that you should round the value before splitting it, as shown by chux in his answer.)

For many plausible money values, you could probably use cents = amount - (int64_t)amount or something similar — but the better way to do it is with modf().

(Beware: there's also the fmod() function for calculating the modulus of two double values, the floating point analogue of the % operator.)

Also, do not use float for real monetary calculations — at least, not accounting style calculations. You can more or less represent values up to $99,999.99 reliably to 2 decimal places, but anything much beyond that is unreliable. (In one casual test, adding 0.01 to 262144.00 didn't change the result.) If you must use floating point numbers, then use double.

Answer 2

Round to the nearest cent.
Break into whole and fractional parts.
Print

 float amount = 25.50;

 amount = roundf(amount*100.0f)/100.0f;
 float dollars, cents;
 cents = modff(amount, &dollars);
 printf("%.0f Dollars  %.0f Cents\n", dollars, cents*100);

Rounding is important to deal with amounts like 1.996 printing as "1 Dollars 100 cents".

modf() nicely separates the fraction from the whole number parts of a floating point number .
Use modf() for double.

Jonathan Leffler is on the money about using double vs. float concerning currency.

Answer 3

When you deal with money with numbers, even if it's just the background of your fun, you really shouldn't have any floating point data type involved. You should use all integers. Not even mention the unreasonability of using float to represent it.

For example:

struct money 
{ 
    long dollars, cents;
}

but normally, money is handled directly from the minimal unit:

typedef money long;

and represent money from cents. When you want dollars:

long dollar_part(money m)
{
    return m / 100;
}

cents:

long cent_part(money m)
{
    return m % 100;
}

so:

money cash;
/* ... */
printf("$ld dollars, $2ld cents", dollar_part(cash), cent_part(cash));

Answer 4

Once you use the mod operator and get the 25, just subtract that value from the original to get the remaining cent value as a float. Multiply by 100 to get the decimal.

Example: 25.50

Methodology:

1. Get the first part of the number (i.e. 25)
2. 25.50-25 = .50
3. .50 * 100 = 50 -> store in variable for later display